Files
nexus/wiki/sources/Multi-Agent-System-Reliability.md

47 lines
2.7 KiB
Markdown
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
---
title: "Multi-Agent System Reliability"
type: source
tags: [multi-agent, reliability, architecture, llm]
date: 2026-04-13
---
## Source File
- raw/AI/Multi-Agent System Reliability.md
## Summary
- 核心主题4种架构模式提升多智能体系统可靠性
- 问题域LLM 本身不可靠(幻觉、逻辑谬误、上下文漂移),多智能体拓扑会将错误传播至系统失效
- 方法/机制Hierarchy层级、Consensus共识、Adversarial Debate对抗辩论、Knock-out淘汰制
- 结论/价值:停止将 LLM 视为"魔法聊天机器人",应视为分布式系统中不可靠组件,需约束、验证、淘汰、挑战
## Key Claims
- LLM 不能被拟人化:它不受生物需求驱动,无法真正"害怕"或"渴望",仅模拟情感
- Hierarchy 模式Supervisor 做计划→分解任务→分配给 Worker→Validator 验证;依赖图强制协作
- Consensus 模式3个模型同时独立处理同一任务选多数票结果同类幻觉概率从20%降至0.8%
- Adversarial Debate 模式:一个生成器提议,一个批评者攻击,一个裁判裁决;防止 Sycophancy阿谀奉承
- Knock-out 模式:多个 Agent 执行任务,最差者淘汰;将 LLM 视为"cattle"而非"pet"
## Key Quotes
> "Stop treating LLMs like magic chatbots. Start treating them like unreliable components in a distributed system." — Alex Ewerlöf
> "We don't need AI that 'cares.' We need AI that is constrained, verified, pruned, and challenged."
## Key Concepts
- [[Multi Agent Hierarchy]]层级模式Supervisor 规划 + Worker 执行 + Validator 验证
- [[Multi Agent Consensus]]共识模式多数投票降低幻觉概率3个模型相同谎言概率降至0.8%
- [[Multi Agent Adversarial Debate]]:对抗辩论模式,防止 Sycophancy真理越辩越明
- [[Multi Agent Knock out]]:淘汰制模式,适应度函数评估,不合格 Agent 直接淘汰
- [[LLM-可靠性工程]]:将 SRE 原则应用于 LLM 系统,视 LLM 为不可靠组件
- [[Sycophancy]]:模型阿谀倾向,用威胁逼迫时可能撒谎以取悦用户
## Key Entities
- [[Alex Ewerlof]]作者资深工程师27年经验SRE 背景2023年起专注 LLM
- [[遗传算法]]GAKnock-out 模式借鉴的经典 ML 方法
## Connections
- [[Multi Agent Hierarchy]] ← 人类组织 ← [[Multi-Agent-System-Reliability]]
- [[Multi Agent Consensus]] ← 民主投票 ← [[Multi-Agent-System-Reliability]]
- [[Multi Agent Adversarial Debate]] ← 法庭对抗 ← [[Multi-Agent-System-Reliability]]
- [[Multi Agent Knock out]] ← 适者生存 ← [[Multi-Agent-System-Reliability]]
- [[LLM]] ← 不可靠组件 ← [[Multi-Agent-System-Reliability]]